The Interagency Workgroup on
Multiple Chemical Sensitivity

August 24, 1998

Predecisional Draft

Use of Biomarkers in Studying MCS

Environmental Control Units

Survey Instruments

Table of Contents

IV. Potential Tools for Future Research Studies

Several tools have been suggested as potentially helpful for a better understanding of MCS. This section will focus on biomarkers and environmental control units as two potential tools that may help clarify the nature of MCS.

Use of Biomarkers in Studying MCS

The use of biomarkers in environmental health was described in a series of publications issued by the Board of Environmental Studies in Toxicology of the National Research Council (NRC, 1989a; NRC, 1989b; NRC, 1992a; NRC, 1992b). Biomarkers were defined as "[i]ndicators of events in biological systems or samples" (NRC, 1987) and were further described as "[t]ools that can be used to clarify the relationship, if any, between exposure to a xenobiotic substance and disease" (NRC, 1989b). (The term xenobiotic denotes a chemical substance that is foreign and perhaps harmful to living organisms.) The NRC (1987) classified biomarkers into three categories based on their relation to the exposure-disease continuum. Biomarkers of exposure were defined as the identification of an exogenous substance within the biologic system, the interactive product between a xenobiotic compound and the endogenous components, or other events in the biologic system related to exposure. Biomarkers of effect were defined as any changes that are qualitatively or quantitatively predictive of health impairment or potential impairment resulting from exposure. Biomarkers of susceptibility were defined as indicators that the health of an organism is especially sensitive to the challenge of exposure to a xenobiotic compound.

In practice, most biomarkers are determined by laboratory tests or functional procedures such as spirometry, tactile threshold, or functional brain imaging. Therefore, many of the same considerations that apply to diagnostic tests and procedures in clinical settings also apply to measuring biomarkers in research applications. Clinical applications are discussed in a separate section, while this section focuses on research.

The use of biomarkers has the potential to help elucidate mechanisms of biologic responses to environmental exposures and, therefore, to identify (or substantially exclude) mechanisms responsible for MCS. Biomarkers of susceptibility may be especially valuable probes for identifying persons at risk for MCS. However, to date, the biomarkers yielding the most useful public health information have been biomarkers of exposure. These include advanced physico-chemical methods that can measure low-level xenobiotics in serum and urine from exposure to: toxic metals (Brody et al., 1994); volatile organic compounds (Ashley et al., 1994); environmental tobacco smoke (Pirkle et al., 1996); pesticides (Hill et al., 1995); aromatic compounds such as dioxins and polychlorinated biphenyls (Pirkle et al., 1995); and other chemical pollutants.

The potential usefulness of biomarker measurements is undermined by either technical or epidemiologic defects (Vineis et al., 1993). Technically, the value of a biomarker can be no greater than the validity of the measurement instrument. Imprecision (poor reproducibility of the measurement) and inaccuracy (significant bias between the measurement result and the true value of the biomarker) can lead to erroneous conclusions in either of two directions: genuine differences can go undetected, or artifactual differences can be created where none really exist (Vineis et al., 1993). The failure to assess and document imprecision and inaccuracy is a frequent shortcoming of research reports addressing biomarkers and MCS, making such findings of little use in public health practice.

The proper use of biomarker tests has been addressed by CDC and ATSDR in a series of reports from workshops and research activities directed at public health investigations at Superfund sites (Hutchinson et al., 1992; Straight et al., 1994; Metcalf et al., 1994; Vogt et al., 1993). These reports recommend that laboratory tests be performed only by laboratories certified under the Clinical Laboratory Improvement Act (CLIA) (Bachner and Hamlin, 1993a; Bachner and Hamlin, 1993b) or engaged in clinical research activities with multicenter quality-assurance programs (Schenker et al., 1993). Tests performed outside these guidelines are often unreliable. The same points apply to biomarkers measured by procedures such as brain scans and neurobehavioral tests, which are often even more difficult to standardize and interpret than laboratory tests (Straight et al., 1995).

Epidemiologic considerations in using biomarkers in public health research are as important as the validity of the measurements. Case definitions of diseases must be specified with sufficient detail and clarity to enable other investigators to reproduce the study. The design must be sound, the investigators and participants properly blinded, the controls suitably selected, and the power of the study must be sufficient to detect differences between the range of normal biologic variation and the expected impact of exposure-related effects. Biomarkers of effect and susceptibility are subject to many confounding physiologic variables such as diurnal variation, stress, nutritional status, concurrent illness, and other contributors to biologic variability, which should be considered in the study design. Although these considerations are true for any public health investigation employing biomarkers, they deserve special emphasis with respect to studies involving MCS because of the lack of established criteria for a case definition. Of the biomarkers reported to have some association with MCS, virtually none has been tested in blinded studies with well-defined populations using methods documented for precision and accuracy.

The costs associated with proper study design for quality assurance are substantial, but researchers and funding centers must accept these expenses as part of the price of sound public health science. The necessity to support quality assurance has been acknowledged in other public health activities, such as the National Health and Nutrition Examination Survey and the Multicenter AIDS Cohort Study (Schenker et al., 1993), but it has been less well recognized in many environmental health settings, particularly those involving MCS. Fortunately, pilot protocols have been organized to standardize laboratory tests for multi-center environmental health studies (Vogt et al., 1990), and the first multicenter study employing them to examine immune biomarkers in MCS cohorts has been initiated. The value of biomarkers in elucidating mechanisms of MCS should become more apparent as the results from such well-designed studies are reported.

Top of Page.

Environmental Control Units

The use of an environmental control unit (ECU) has been suggested by many investigators as a potential means of clarifying MCS mechanisms. An ECU is constructed of nonsynthetic materials that do not release vapors (known as "off gassing") and is ventilated through a filtered, positive-pressure air supply. Furniture within the chamber is constructed of natural, nonsynthetic materials. Attendants and others working in the ECU must not wear dry-cleaned clothing and scented cosmetics. The typical challenge protocol involves placing the patient in the ECU to eliminate exogenous sources of contamination. After a period of time (4-7 days or longer), during which time deadaptation or "unmasking" occurs, the patient is challenged with various potential triggers of symptoms, and reactions are noted (Ashford and Miller, 1991).

An ECU, depending on its use, can be a form of exposure chamber, which is an enclosed space specifically designed for the conduct of inhalation toxicology studies. Exposure chambers are used in animal toxicology studies and some human investigations. They are designed to control precisely the ventilation, temperature, and humidity of the chamber's interior air, and they contain equipment to carefully deliver measured concentrations of chemicals (e.g., solvent vapors into the chamber. Many protocols have been reported for chambers and have been used for diagnosis, treatment, and research (Selner, 1996).

Many complexities are associated with use of ECUs. Care must be provided around the clock (often a week or longer), along with special clothes, foods, reading materials, and other supplies. The costs of operation and maintenance, therefore, are substantial. Many factors can affect how subjects respond in ECUs. They include temperature range, intensity and type of light in the chamber, and airflow rate. To accurately test chemicals at or near the odor threshold, the use of masking agents has been recommended, although some researchers feel that experiments should begin with a natural, unmasked exposure. Finding an agent that does not cause a response can be difficult (Selner and Staudenmayer, 1995).

Two other points merit comment relevant to the use of ECUs in human subjects research on chemical sensitivity. First is the ethical question of exposing persons to substances in ECUs that may cause them to suffer symptoms of ill health. This kind of ethical concern is considered by Institutional Review Boards, which must review and approve any human research study protocol. Another consideration is the degree to which an ECU represents an unnatural environment to persons who are research subjects. If a person responds positively to being in an ECU, would the experience possibly increase later self-isolation if the patient attempts to re-establish the conditions found in an ECU?

Although these complexities are difficult to overcome, investigators in the United States who have a great deal of experience are using ECUs in their research. Few units are currently in use in the United States, however, primarily because they are very expensive to operate and there are questions concerning their efficacy in routine diagnosis and treatment.

For research purposes, ECUs may offer the possibility of learning whether many of the etiologies and mechanisms suggested for MCS can be validated. Some scientists and physicians believe that valuable information could be gained by the proper use of an ECU (AOEC, 1992; Ashford and Miller, 1991; Miller, 1994; NRC, 1992; Selner, 1996).

Top of Page.

Survey Instruments

Questionnaires are one of the most useful tools in epidemiologic investigations. With careful definition of terms, questionnaires can be both an efficient and effective means to collect information about MCS, even in the absence of a consensus case definition. An effective questionnaire will endeavor to define its terms operationally. Questionnaires that have been used in MCS research can be grouped into three subject categories: symptoms, chemicals, and lifestyle changes.

MCS symptoms questionnaires query the participant about symptoms related to specific organs (e.g., itchy, burning, red, or watery eyes; difficulty focusing) and to constitutional symptoms (e.g., fever and fatigue). Good symptom questionnaires seek to define the symptom, the number of organ systems involved, and the degree of symptom severity.

Questions about the chemicals ask about the substance implicated in the first episode of sensitivity. Information about the number of chemical classes that trigger reactions and the time sequence of the exposure and symptoms is sought. Questions about whether "spreading" of triggering to new chemicals has occurred are also asked.

Assessment of changes in lifestyle are considered an important way to assess the severity of MCS. Questions about changes in diet, working, and shopping are asked. Changes in home furnishing and clothing may also be queried.

 

Top of Page.